exp-no0-hack3
This model is a fine-tuned version of google-bert/bert-base-multilingual-cased on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.1971
Model description
More information needed
Intended uses & limitations
More information needed
Training and evaluation data
More information needed
Training procedure
Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 1e-05
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 10
Training results
Training Loss | Epoch | Step | Validation Loss |
---|---|---|---|
0.4397 | 1.0 | 443 | 0.4302 |
0.2741 | 2.0 | 886 | 0.2767 |
0.4316 | 3.0 | 1329 | 0.2856 |
0.3033 | 4.0 | 1772 | 0.2203 |
0.0856 | 5.0 | 2215 | 0.2243 |
0.0484 | 6.0 | 2658 | 0.2258 |
0.3601 | 7.0 | 3101 | 0.2291 |
0.2423 | 8.0 | 3544 | 0.1641 |
0.0145 | 9.0 | 3987 | 0.2100 |
0.0884 | 10.0 | 4430 | 0.1971 |
Framework versions
- Transformers 4.40.1
- Pytorch 2.2.1+cu121
- Datasets 2.19.1
- Tokenizers 0.19.1
- Downloads last month
- 2